Effect of temperature-photoperiod interaction on growth and winter bud development in Norway spruce (Picea Abies)

(1)

Master’s Thesis 2020 60 ECTS

Faculty of Biosciences

Effect of temperature-photoperiod interaction on growth and winter bud development in Norway spruce (Picea Abies)

Katharina Therese Hobrak

Master in Biology

(2)

(3)

1

The Norwegian University of Life Sciences

Norges miljø- og biovitenskapelige universitet

Master thesis

Effect of temperature-photoperiod interaction on growth and winter bud development in Norway spruce (Picea Abies)

Katharina Therese Hobrak

Department of Plant Science Ås, 2020

The Norwegian University of Life Sciences

P.O Box 5003, 1432 Ås, Norway

(4)

2

Abstract

Woody species growing in temperate and boreal regions, like Norway spruce (Picea abies) have to enter dormancy to survive winter and freezing temperatures, while simultaneously maximizing their growing period. Dormancy is a temporary inability to resume growth, even though the plant experiences favourable growth conditions. Dormancy is usually initiated by growth cessation and bud set. Many species, including Norway spruce, use the daylength, also called photoperiod, as a signal to induce seasonal life events. Such plants respond to

photoperiods shorter than a certain daylength, called the critical daylength with growth cessation, bud set and further dormancy development in the autumn. In such plants

photoperiods longer than a certain daylength, called the critical daylength, sustain growth. In some species, a long photoperiod is also required for bud break and re-growth. The length of the critical daylength varies both between species, but especially between provenances.

Provenances are local populations that have adapted to local climatic conditions and daylength. Provenances from higher latitudes usually have a longer critical daylength for growth than those from lower latitudes. Temperature is also an important environmental factor affecting both dormancy and regrowth. With temperature both increasing in the past and predicted to increase further in the future, it is highly relevant to study the effect of temperature on the phenology in plants. Several studies have been conducted on plants responses to temperature and short days (SD) and contradictory results have been found:

Studies on several species conducted in growth chambers have found that bud set occurs earlier when the plants are exposed to warmer compared to colder temperatures, while a number of field studies have found opposite results, with colder temperatures resulting in faster bud formation. In growth chamber studies, the plants have commonly been placed directly to SD shorter than the critical daylength for growth under constant temperature or alternating day and night temperature involving rapid changes. Such daylength and

temperature regimes may possibly stress the plants since daylength and temperature changes are gradual in nature.

The aim of this MSc thesis has been to study the effect of temperature on seedlings from the Halden (59°N) and Rana (66°N) provenances (both from Norway) of Norway spruce exposed to different bud set-inducing SD conditions. Specifically, it was tested whether the growth cessation and bud set response to temperature differed in plants exposed to gradually decreasing daylengths (24 h to 12 h) and plants exposed to constant SD conditions of 12 h photoperiod. The temperature regimes were either a) constant temperature of 12, 18 or 24°C

(5)

3

under SD of 12 h and LD of 24 h photoperiod for comparison, or b) 12 or 18°C under gradually decreasing daylength, or c) gradually changing, alternating day and night temperatures of 18/12°C or 24/18°C day/night temperature in combination with gradually decreasing daylengths. In addition, the effect of the different temperatures on various other growth parameters was studied. Afterwards, all plants were re-transferred to LD and 18°C to study the after-effect of the temperature and daylength treatments on bud break and re- growth.

The results showed that both the plants given decreasing daylengths (2 h per week down to 12 h photoperiod) and plants exposed to SD of a 12 h constant photoperiod, ceased growth and showed earlier bud set when grown at warmer temperature. The plants given decreasing daylengths had a delayed bud set response, compared to the constant 12 h SD, and plants from the northern provenance (Rana) showed faster bud set than the plants from the more southern provenance (Halden). In addition, more growth was generally observed (for most growth parameters) when the plants were kept at warmer temperatures, and plants from Halden generally grew more than those from Rana. Under constant daylengths, the plants that were exposed to 24°C did not differ from those at 18°C as much as plants at 12°C differed from those at 18°C, both with respect to growth and bud set. Furthermore, in plants exposed to alternating, gradually changing day and night temperature and decreasing daylength, bud set was more rapid under 24/18°C day/night than 18/12°C, with the Rana-plants showing earlier bud set than the Halden-plants. In both daylength treatments (combined with constant

temperatures), bud break after subsequent transfer to LD and 18°C was the fastest in the plants that had been exposed to 12°C, indicating less deep dormancy in these plants compared to those from the higher temperature regimes. However, re-growth in plants from 12°C was only faster in the plants that were exposed to the decreasing daylengths.

In conclusion, the response to the different tested temperature regimes was similar with earlier bud set at the highest temperature both under gradually decreasing daylengths and constant SD of 12 h photoperiod. Thus, the specific daylength regimes tested did not affect the overall bud set response to temperature.

(6)

4

Sammendrag

Trearter som vokser i tempererte og boreale strøk slik som gran (Picea abies) må gå i vinterhvile for å overleve vinteren og minusgrader, mens de samtidig skal maksimere sin vekstperiode. Vinterhvile er en midlertidig manglende evne til å gjenoppta vekst, selv om planten opplever gunstige vekstvilkår. Vinterhvile er vanligvis innledet av vekstavslutning og knoppsetting. Mange arter, inkludert gran, bruker daglengden, også kalt fotoperiode, som et signal for å indusere sesongavhengige livshendelser. Disse plantene responderer på

fotoperioder kortere enn en viss daglengde, kalt kritisk daglengde, med vekstavslutning, knoppsetting og videre hvileutvikling på høsten. Er fotoperioden lengre enn en viss daglengde, opprettholdes veksten. I noen arter er en lang fotoperiode også nødvendig for knoppbrytning og for å gjenoppta vekst. Lengden av den kritiske daglengden varier mellom arter, men særlig mellom provenienser. Provenienser er lokale populasjoner som har tilpasset seg til lokale klimatiske forhold og daglengder. Provenienser fra høyere breddegrader har vanligvis en lengre kritisk daglengde for vekst enn de fra lavere breddegrader. Temperatur er også en viktig miljøfaktor som påvirker vinterhvile og gjenvekst. Siden temperaturer både har økt i fortiden og forutsees å øke mer i fremtiden, er det høyst relevant å studere effekten av temperatur på fenologien i planter. Flere studier har blitt utført på planters respons på temperaturer og korte dager (KD) og her har motstridende resultater blitt funnet: Studier av flere arter utført i vekstkamre har funnet at knoppsetting skjer tidligere hvis planten utsettes for varmere, sammenlignet med kaldere temperaturer, mens noen feltstudier har funnet motsatte resultater, det vil si at kaldere temperaturer resulterer i raskere knoppdannelse. I vekstkammerstudier har planter vanligvis blitt plassert direkte til KD kortere enn den kritiske daglengden for vekst, under konstante temperaturer eller vekslende dag- og natt-temperaturer med raske endringer. Slike daglengde- og temperaturregimer stresser muligens plantene siden daglengden og temperaturen endrer seg gradvis i naturen.

Målet med denne masteroppgaven har vært å studere effekten av temperatur på frøplanter av granprovenienser fra Halden (59°N) og Rana (66°N) (begge fra Norge) eksponert for

forskjellige knopp-induserende KD-forhold. Nærmere bestemt ble det testet om

vekstavslutnings- og knoppsettingsrespons på temperatur varier mellom planter eksponert for gradvis minkende daglengder (24 t til 12 t) og planter eksponert for konstante KD-forhold med 12 t fotoperiode. Temperaturregimet var enten a) konstante temperaturer ved 12, 18 eller 24°C under KD med 12 t fotoperiode og LD med 24 t fotoperiode til sammenligning, eller b) 12 eller 18°C under gradvis minkende daglengder, eller c) gradvis endrende, vekslende dag-

(7)

5

og natt-temperaturer ved 18/12°C eller 24/18°C dag/natt-temperaturer i kombinasjon med gradvis minkende daglengder. I tillegg ble effekten av forskjellige temperaturer på diverse andre vekstparametere studert. Etterpå ble alle plantene tilbakeført til LD og 18°C for å studere etter-effekten av temperatur- og daglengdebehandling på knoppbrytning og gjenvekst.

Resultatene viste at både plantene som ble gitt minkende daglengder (2 t i uka ned til 12 t fotoperiode) og plantene som ble eksponert for KD med 12 t fotoperiode, avsluttet vekst og dannet knopper tidligere når de fikk varmere temperaturer. Plantene som ble gitt minkende daglengder hadde en forsinket knoppsettingsrespons, sammenlignet med de som fikk konstant 12 t KD. Planter fra den nordlige proveniensen (Rana) viste raskere knoppsetting enn plantene fra den mer sørlige proveniensen (Halden). I tillegg ble det observert mer vekst for de fleste vekstparameterne når plantene fikk varmere temperaturer, og plantene fra Halden vokste mer enn de fra Rana. Under konstante daglengder var forskjellene i vekst og knoppsetting mellom plantene som ble eksponert for 24°C og 18°C mindre enn mellom plantene fra 12°C og 18°C.

I plantene som ble eksponert for forskjellige, gradvis skiftende dag- og natt-temperaturer og minkende daglengde, var knoppsettingen raskere under 24/18°C dag/natt-temperatur enn 18/12°C, og Rana-plantene dannet knopper tidligere enn plantene fra Halden. I begge daglengdebehandlinger (kombinert med konstante temperaturer) var knoppbrytning etter tilbakeføring til LD og 18°C raskest i plantene som ble eksponert for 12°C, noe som indikerer en vinterhvile som er mindre dyp sammenlignet med planter fra høyere temperaturregimer.

Gjenvekst i plantene fra 12°C derimot, var bare raskere i plantene fra minkende daglengder.

Samlet sett viste plantene lignende respons på de forskjellige temperaturregimene med tidligere knoppsetting ved den høyeste temperaturen både under gradvis minkende daglengde og konstant KD med 12 t fotoperiode. De spesifikke daglengderegimene som ble testet, påvirket derved ikke den generelle knoppsettingsresponsen på temperatur.

(8)

6

Acknowledgements

First, I would like to thank my main supervisor Jorunn E. Olsen for the dedicated guidance, both throughout the experimental phase and through the writing of the master thesis. Then I would like to thank my co-supervisor YeonKyeong Lee with the help and guidance in microscopy of the buds. I would like to thank Marit Siira for helping with registrations, and for watering and taking care of the plants during the experimental treatments and also thanks to SKP for taking care of all the technical set up during the experiments. Thanks to Marcos Viejo for letting me use his data. And thanks to Christian Strømme for helping with the statistical analyses in R. And lastly a special thanks to Daniel Flaten Sunde for technical support in Word.

Abbreviations

SD Short day*

LD Long day*

R Red light

FR Far red light

PHY Phytochrome gen

B Blue light

GA Gibberellin

ABA Abscisic acid

m.a.s.l. Metres above sea level RH Relative air humidity

VPD Water vapour pressure deficit EC Electrical conductivity PBS Sodium phosphate buffer ANOVA Analysis of variance

(9)

7 Clmm Cumulative link mixed model

AICc Akaike information criterion corrected

* The length of long and short day described in the introduction varies and depends on the study that is being discussed, while long and short day in this study generally refers to 24 h and 12 h respectively.

Keywords

Bud break, bud set, growth, growth cessation, Norway spruce, photoperiod, provenances, temperature,

(10)

8

Introduction

Preface

The daylength varies both within a year and across the globe and with increasing distance from the equator, the cycles of daylengths are more extreme. These daylight cycles contribute to the seasonal temperature variations. In temperate and boreal regions, the winters are too cold for plants to be able to grow. Not being able to grow all year round, makes it important to utilize the growing period effectively, while simultaneously developing frost tolerance in time before the winter and freezing temperatures. The temperature fluctuates within seasons and between years and is therefore not a reliable indication of the time of the year. Daylength on the other hand, cycles annually and is a more accurate indicator of the time of the year.

Therefore many lifecycle events in plants that depend on a certain season respond to

daylength signals, but other environmental factors like temperature and light quality are also important signals (Olsen, 2010; Taiz et al., 2015).

Photoperiodism

The specific daylength experienced by a plant, is referred to as photoperiod, while the ability to detect the photoperiod and respond to it is called photoperiodism (Garner & Allard, 1923;

Taiz et al., 2015). Garner and Allard were pioneers in studying how plants use photoperiod to control flowering (Garner & Allard, 1922; Taiz et al., 2015). Since then photoperiodism has been extensively studied, most often with respect to flowering, but plants also use

photoperiods to control other life-cycle events that depend on a specific season, like

dormancy in the autumn or regrowth in the spring (Clapham et al., 1998; Nitsch, 1957; Olsen, 2010; Taiz et al., 2015). The response to photoperiods largely depends on a critical daylength.

With respect to flowering, plants are usually divided into short-day and long-day plants, depending on whether the induction of flowering depends on a photoperiod that is shorter (short-day plant) or longer (long day plant), then a critical daylength (Jackson, 2009; Taiz et al., 2015). In a wide range of woody plants from the temperate and boreal zone, growth is controlled in a similar way; short days (SD) induces growth cessation, and long days (LD) promotes growth. The main focus from here on out will be photoperiodism in growth control.

Plants can also be divided into dark-dominant and light-dominant plants, depending on whether the night length (dark-dominant) or the length of the day and light quality (light- dominant) is the most important factor. In woody plant species, often populations from higher latitudes tend to be more light-dominant, while plants from lower latitudes tend to be more

(14)

12

dark-dominant (Clapham et al., 1998; Olsen, 2010). Light quality will be further addressed below.

Plants keep track of time with the circadian clock, which results in an internal daily cycle that is being set both by environmental factors but is also able to keep the rhythm by endogenous factors, at least for a while after transfer to constant conditions. The circadian clock is among others a part of the mechanisms behind photoperiodism (Eriksson & Millar, 2003; Taiz et al., 2015). Phytochromes are important photoreceptors, known to be involved in photoperiodism.

In flowering control cryptochromes are also known to be important, in woody species this is yet to be confirmed (Eriksson & Millar, 2003; Olsen, 2010). Phytochromes mostly respond to wavelengths in the red (R)/far red (FR) spectrum. There are several types of phytochrome genes, e.g. the angiosperm model plant Arabidopsis thaliana has PHYA-PHYE, but not all angiosperm species have all 5, e.g. Populus species appear to have 3 types; a PHYA gene, and 2 variants of PHYB. The gymnosperm woody species Norway spruce (Picea abies) and Scots pine (Pinus sylvestris) have PHYN, PHYO and PHYP, which are homologs to PHYA and PHYC and PHYB, PHYD and PHYE respectively (Clapham et al., 1999; Olsen et al., 1997;

Opseth et al., 2016; Taiz et al., 2015). Olsen et al. (1997) showed that overexpression of PHYA in the woody hybrid aspen (Populus tremula x tremuloides), changed the critical daylength, and the plant continued to grow under SD. PHYA therefore seems to be important photoperiodic control in growth.

Besides photoperiod, light quality is also important in growth control. As mentioned above, light quality is relatively more important at higher latitudes, compared to lower latitudes (Mølmann et al., 2006; Olsen, 2010). In a study of Norway spruce FR was more effective in maintaining growth than R but a 1:1 ratio of R:FR was the most effective in this respect. Blue light (B) did not stop growth cessation, but delayed bud formation (Mølmann et al., 2006).

Phenology

Phenology is the study of the life cycle events a plant goes through in a year or a lifetime and the seasonal environmental factors that induce these changes (Körner, 2012; Njoku, 2014;

Rathcke & Lacey, 1985). These changes can be dormancy release, bud burst and leafing in the spring, flowering and fruiting, or growth cessation, bud formation and dormancy induction and development as well as leaf senescence and abscission during autumn. The phenology of

(15)

13

a plant is important for the optimal utilization of resources, e.g. maximizing growth, while avoiding the risk of freezing in the winter (Chuine, 2010; Delpierre et al., 2016).

Dormancy

Woody plants grow above ground for many years, even up to thousands of years for some species, and in temperate and boreal regions they therefore depend on surviving the winter.

Winters are often both too cold and too dark for growth and freezing damage is a potential risk. Plants in such areas therefore cease growth, form buds, enter dormancy and get frost tolerant (Olsen, 2010; Rohde et al., 2000; Welling & Palva, 2006).

Lang et al. (1987) defined dormancy as “a temporary suspension of visible growth of any plant structure containing a meristem”. Furthermore three different types of dormancy were recognized depending on the cause of growth suspension: Ecodormancy, where unfavourable environmental conditions cause lack of growth. Paradormancy and endodormancy are growth suspension caused by endogenous factors; in paradormancy the signal originates in another organ, while in endodormancy growth suspension is caused by signals within the affected organ. Before Lang et al. (1987) defined these three types of dormancy, a range of different terms were used (Lang et al. (1987) recognized 54 different terms used in literature) to describe the three types of dormancy, leading to confusion. Examples of terms used among many others are: Imposed dormancy and quiescence (ecodormancy), summer-dormancy and correlative inhibition (paradormancy) and winter-dormancy and rest (endodormancy). Junttila (1988) commented that the Lang et al. (1987) definition was to wide and included growth suspension caused by unfavourable conditions, which was not considered as dormancy earlier. To make the definition easier to understand Rohde and Bhalerao (2007) proposed a new definition of dormancy that is now widely approved: “The inability to initiate growth from meristems (and other organs and cells with the capacity to resume growth) under favourable conditions”, presupposing that there is an ability for re-growth. By this definition, ecodormancy is not considered a type of dormancy. Further on, the terms from Rohde and Bhalerao (2007)will be used.

Apical growth cessation is often induced by the short photoperiods of the autumn and is usually the first event to happen, and important to enable dormancy (Nitsch, 1957; Olsen, 2010; Rohde & Bhalerao, 2007; Wareing, 1956). Short photoperiods mostly induce growth cessation in species and developmental stages that exhibit a free growth pattern, i.e. plants

(16)

14

that form leaf initials simultaneously with internode elongation (Olsen, 2010). Species and developmental stages that exhibit a fixed growth pattern i.e. plants where formation of leaf initials occurs separate from internode elongation, will usually stop growing while there still is a long photoperiod. Although this varies between species, a free growth pattern is typically a juvenile trait, and older plants more often have a fixed growth pattern (Olsen et al., 2004).

Also, the formation of buds is often induced by short photoperiods (Olsen, 2010; Wareing, 1956). It usually takes a few weeks from induction, until the first bud can be seen, depending on the species and its latitudinal origin (Olsen, 2010). Buds contain leaf primordia, that are protected by bud scales (Delpierre et al., 2016; Lee et al., 2017). Although a short photoperiod is the most important signal, temperature also affects dormancy (Junttila et al., 2003; Kalcsits et al., 2009; Strømme et al., 2017; Tanino et al., 2010). Effects of temperature is further described below. Together with growth cessation and bud formation, plants start to build up freezing tolerance, which is vital for surviving winter. Cold acclimation is also induced by short photoperiods, this is also the case for plants with a fixed growth pattern. After induction by a short photoperiod, both the exposure to cold temperatures and prolonged short

photoperiods increases the frost tolerance substantially (Howe et al., 2003; Olsen, 2010;

Welling & Palva, 2006).

In the beginning of growth cessation and bud formation, the plants normally have not entered dormancy yet (Junttila et al., 2003; Rohde & Bhalerao, 2007). As time progresses, the plant will gradually enter a dormant state. It takes about 1-3 weeks to induce growth cessation, this varies both between species and within species, while it takes another 2-3 weeks to induce dormancy.

The phytohormones that are involved in dormancy regulation are still not fully identified.

However, down-regulation of gibberellin (GA) has been shown to be involved in growth cessation but does not play a further role in dormancy development (Mølmann et al., 2005).

Ethylene is likely to be involved in bud development, so is abscisic acid (ABA) (Olsen, 2010). It seems that ABA is involved in the differentiation of bud scales and leaf primordia as well as cold acclimation (Basler & Körner, 2014; Olsen, 2010).

In the spring, as the length of the photoperiod and temperature increases, plants exit dormancy and break buds and resume growth (Basler & Körner, 2014; Olsen, 2010; Welling & Palva, 2006). In many plants species dormancy release requires chilling, which is a certain time period, below a certain temperature (Basler & Körner, 2014; Junttila et al., 2003; Rohde &

(17)

15

Bhalerao, 2007). Chilling does not promote growth; it only restores the ability to grow (Rohde

& Bhalerao, 2007). The temperature required for chilling is not fully understood, but it is believed that temperatures just above freezing, are most effective (Basler & Körner, 2014).

Spring phenology with bud burst and start of growth is complex and very species-dependent (Basler & Körner, 2014; Roberts et al., 2015). Temperature, photoperiod and chilling are all factors that affect growth start and bud burst, but to varying degrees in different species (Basler & Körner, 2014; Heide, 2003). Both autumn temperature and spring temperature can affect the spring phenology. Autumn temperature can affect the depth of dormancy, and thereby the requirement for chilling. In some species, exposure to long photoperiods for some time, replaces the requirement for chilling (Heide, 2003). Increased temperature in spring accelerates bud burst in many species (Basler & Körner, 2014; Heide, 2003).

Provenances

As mentioned above, seasonal temperature and light fluctuations depend on latitude. Many species have a distribution range across many latitudes and will therefore face different climatic conditions along a latitudinal gradient. Populations therefore adapt to local

conditions, and these populations are called ecotypes or provenances (Heide, 1974; Körner, 2012; Vaartaja, 1959). While temperatures generally decrease at higher latitudes, the

photoperiod has a more extreme cycle with longer days in summer and shorter days in winter, compared to lower latitudes. The critical daylength that plants respond to is therefore among the local adaptations (Vaartaja, 1959). Usually the critical daylength increases with increasing latitude, both because the plants need to enter dormancy earlier and because the days usually are longer when the plants start the induction towards dormancy (Heide, 1974; Olsen et al., 1997; Vaartaja, 1959).The altitude the plants grow at can also alter the critical daylength, since temperature decreases with altitude. As mentioned earlier, plants from higher latitudes are also increasingly dependent on the light quality rather than just the critical daylength (Olsen, 2010).

Norway spruce – Picea abies

Norway spruce (Picea abies (L.) Karst.), is a coniferous tree species, which is one of the main species in temperate and boreal forests of Europe (Jansson et al., 2013). The species has its natural distribution in large parts of Scandinavia and eastern Europe, in mountains of central and south eastern Europe, where it can grow at altitudes higher than 2300 m.a.s.l., and has

(18)

16

also been planted outside its natural range, like in Denmark, northern Germany, Scotland, France, Iceland and parts of North America. Norway spruce is an important tree in forestry in Europe. Its wood is being used for several different purposes, such as paper production and timber construction, and it is commonly used as Christmas tree (Aarnes, 2014; Anderberg, 2004; Jansson et al., 2013). Norway spruce is a climax species and is often one of the

dominating species in its ecosystem in Scandinavian forests. (Anderberg, 2004; Aune, 2019;

Fremstad, 1997; Jansson et al., 2013). It is an evergreen species, and often forms dense forests. The leaves are needle-like and the needles stay on the tree for 8-10 years, although shoots with new needles are produced every year (Aarnes, 2014; Anderberg, 2004). Norway spruce displays apical dominance, giving it the characteristic cone shape. The juvenile stage is relatively long, maturity is reached at about 20 years (Gyllenstrand et al., 2007). Norway spruce can live for about 400 years and can get up to 40-50 m tall (Aarnes, 2014; Vidakovic, 2020)

Norway spruce has a growth cycle common for trees in temperate and boreal forests, where the trees enter dormancy in autumn and exit dormancy in spring (more details about

dormancy described above) (Dormling et al., 1968; Gyllenstrand et al., 2007; Mølmann et al., 2006). Young seedlings of Norway spruce have a free growth pattern, while older specimens have a fixed growth pattern (Gyllenstrand et al., 2007). Norway spruce seems to have a shallow dormancy and a low chilling requirement, at least in young individuals (Dormling et al., 1968; Olsen et al., 2014). If transferred back to long photoperiods after bud formation under short days, buds will burst rapidly, within a few weeks (Olsen et al., 2014). The latitudinal range extends from about 41°N to about 72°N, and photoperiodic provenances have evolved (Jansson et al., 2013). The critical daylength for provenances at around 55°N is about 16 h while provenances at 63°N have a critical daylength of about 18 h (Thomas &

Vince-Prue, 1996)

Climate change and how temperature affects the growth cycle

Since 1880 the average global surface temperature has increased with 0.85°C and is predicted to increase even further in the future, and this is most likely due to anthropogenic greenhouse gas emissions (Pachauri et al., 2014). It is predicted that the temperature will continue to increase in the future, though it is uncertain how much, as it largely depends on future greenhouse gas emissions.

(19)

17

As already mentioned briefly above, temperature is also involved in the control of the growth cycle (Olsen, 2010; Olsen & Lee, 2012; Tanino et al., 2010). Therefore, an increase in

temperature will potentially affect the phenology of plants. Several studies have been conducted to figure out how temperature affects growth control in various woody species.

Some woody species from the Rosacea family like apple (Malus pumila) and pear (Pyrus communis) do not respond to short photoperiods, and it seems that cold temperatures (<12°C) induces dormancy (Heide & Prestrud, 2005). Studies on other species have shown contrasting results. Growth chamber studies conducted on several woody species, including Norway spruce, white spruce (Picea glauca), silver birch (Betula pendula), downy birch (Betula pubescens) and hybrid poplar (Populus x spp.) have shown faster bud formation or entered dormancy faster under higher compared to colder temperature treatments (Hamilton et al., 2016; Junttila et al., 2003; Kalcsits et al., 2009; Olsen et al., 2014; Tanino et al., 2010). On the other hand, field research have shown that higher temperature delays bud formation in

Eurasian aspen (Populus tremula) and hybrid poplar (Rohde et al., 2011; Strømme et al., 2015; Strømme et al., 2017; Strømme et al., 2018). It could be noted that studies of Populus species in the field and growth chambers have shown contradictory responses with respect to effect of temperature under short days on the timing of bud set. Zohner and Renner (2019) studied how increased temperature affected 8 woody species, grown in a greenhouse. They increased the temperature in parts of the year, or all year around relative to the ambient temperature, with control plants that received the ambient temperature, and found that increase in temperature during summer and autumn delayed bud set compared to the control, whereas an increase in temperature all year around or during the winter and spring accelerated bud set compared to the control.

The dynamics of temperature in dormancy release and bud break has been studied in greater detail. High temperatures during short day treatment in growth chambers has been shown to deepen dormancy, increase chilling requirement and delay bud burst during long day

treatment, in a range of species like Norway spruce, downy birch, silver birch, black alder (Alnus glutinosa) and Norway Maple (Acer platanoides), indicating that increased

temperature during autumn will delay bud flushing during spring (Dormling et al., 1968;

Heide, 2003; Olsen et al., 2014; Søgaard et al., 2008; Tanino et al., 2010; Westergaard &

Eriksen, 1997). Increased temperatures during spring have shown to cause earlier bud break in different species including sycamore (Acer pseudoplatnaus), European beech (Fagus sylvatica), Norway spruce, sessile oak (Quercus petraea) and Eurasian aspen (Basler &

(20)

18

Körner, 2014; Strømme et al., 2015; Strømme et al., 2018; Strømme et al., 2019). Roberts et al. (2015) analysed the Marsham phenology time series, which was a time course of the first leafing dates in 13 tree species, which was recorded in the period 1753-1947 in Norfolk, England. They found that the winter and spring temperature affected the different species to various degree, and that some species were more affected than others by temperature and showed larger differences between colder and warmer temperature. Also, some species started leafing earlier than others in response to warmer temperatures.

In growth chamber studies, plants are commonly raised under long day treatment and then transferred directly to a short day treatment (Hamilton et al., 2016; Junttila et al., 2003;

Kalcsits et al., 2009; Olsen et al., 2014). It is therefore possible that the sudden change in day length may cause a stress response in the plants, and that plants kept at warmer temperatures respond faster to this, than plants kept at colder temperatures. It is also common to grow the plants under a constant temperature or under different day and night temperatures, keeping each of them constant.

Aims of the study This study aimed to:

• Evaluate the effect of temperature on growth and bud set under short days of 12 h in seedlings of a southern and northern provenance of Norway spruce, (from Halden at 59°N latitude and Rana at 66°N latitude, respectively) and the effect of temperature on growth under long days of 24 h daylength.

• Investigate if growth and bud development in seedlings of these southern and northern Norway spruce provenances respond differently to temperature treatment if they are given a gradual shortening of the daylength, compared to plants transferred to a 12 h day length directly from 24 h daylength.

• Evaluate if alternating day and night temperatures affect the growth and development of these southern and northern Norway spruce provenances differently than constant temperature during gradual decrease in daylength shortening.

(21)

19

Materials and methods

Study species and seed provenances

The study species was Norway spruce (Picea abies (L.) H. Karst.). Seeds from two different provenances were used (figure 1). One provenance was from the seed collection area CØ/1, Halden, Østfold, Norway, which is located at 59 °N (seed lot 980863, Skogfrøverket, Hamar, Norway). The other provenance was from P/1, Rana in Nordland, Norway, located at 66°24’N (seed lot 4145, Skogfrøverket, Hamar, Norway). The letters in CØ/1 and P/1 represent

location and the numbers indicate 0-149 metres above sea level (m.a.s.l) (figure 1).

Figure 1 Map showing the collection areas for forest tree seeds in Norway, including the origin of the 2 provenances Rana and Halden of Norway spruce used (P/1 and CØ/1), used for studying the effect of temperature and daylength treatments in seedlings. The altitudinal zones indicate height above sea level categories, m.a.s.l. = metres above sea level. (Modified from maps by Skogfrøverket and Statens kartverk; http://www.skogfroverket.no/artikkel.cfm?Id_art=10&kanal=3 )

Sowing and pre-growing before the experimental treatments

Seeds were sown either in S-soil (Hasselfors, Örebro, Sweden) (experiment 1a and 1b, Table 1) or a growth peat (Degernes torvstrøfabrikk, Degernes, Norway) and perlite mixture (1.5-6 mm, Agro, Ankara, Turkey), with a ratio of 1:3 perlite to peat (experiment 2 and 3, Table 1). The seeds were sown in pots with a dimension of 5.5 x 4.5 x 4 cm (upper diameter, height, lower diameter). The pots were placed in trays with 12 pots in each. Two seeds were

(22)

20

sown in each pot, to ensure germination of at least one plant per pot, since the germination rate was estimated to be about 60%.

After sowing, the trays were placed in a growth chamber (manufactured by the Norwegian University of Life Sciences), where the seeds had 6 (experiment 2 and 3) and 8 (experiment 1a and 1b) weeks to germinate and grow. In the growth chamber, the temperature was 18 °C, and the relative air humidity (RH) was 76%, which gives a water vapour pressure deficit (VPD) of 0.5 kPa. The plants were given 24 h of light treatment per day called long day (LD).

Of these, 12 h were with 180-200 μmol m^-2 s^-1 irradiance, using metal-halide lamps (Master HPI-T Plus 400 W/64 E40 1SL, Phillips, Amsterdam, Netherlands) and light from

incandescent lamps (mixture of Osram, Munich, Germany and Narva, Brown & Watson International Pty Ltd, Knoxfield, Australia); called full light. During the remaining 12 h light of the diurnal cycle, only incandescent lamps were used, with an irradiance of 8-10 µmol m^-2 s^-1, called day extension light. At full light the ratio between the red (R) and far red (FR) light was adjusted to 1.7 with the incandescent light.

The plants were watered as needed. The plants were fertilized after one week of growth and were thereafter fertilized twice a week. The nutrient solution contained calcium nitrate, ammonium nitrate and Kristalon (Yara, Oslo, Norway), with an electrical conductivity (EC) of 1.5. The plants were watered with nematodes, in the case of insects. Flies were trapped with fly paper.

Treatment for winter bud formation

After 6-8 weeks (as described above) of pre-growing, treatments for bud formation started. In pots with two or more seedlings, excess seedlings were removed using a pair of scissors, keeping the most centred, the straightest and most evenly sized seedlings, leaving one

seedling per pot. Thus, as similar plants as possible were used in each of the experiment. The treatments and the number of plants per treatment and provenance are summarized in Table 1.

The experiments described in Table 1 where conducted in the order indicated by the table.

(23)

21

Table 1. Overview over treatments in experiments studying the effect of temperature and daylength treatments on growth and development in seedlings of Norway spruce.

Experiment Light treatment

Provenance Temperature treatment

Relative humidity¹ VPD² = 0.5 kPa

Length of bud set treatment

Length of bud break treatment

Number³ of plants per treatment/

provenance

1a 12 h

photoperiod (SD)

24 h photoperiod (LD)

Halden 12°C 64% 50 days 28 days 20 plants

18°C 76%

24°C 83%

1b Rana 12°C 64% 25 days

18°C 76%

24°C 83%

2 22 h -12 h

photoperiod, decreased with 2 h a week

Halden Rana 12°C 64% 63 days 35 days 36 plants

18°C 76%

3 18°C day/

12°C night

76% day/

64% night

56 days No bud break treatment 24°C day/

18°C night

83% day/

76% night

1: The relative humidity (RH) was set to the values in the table, but during alternating day and night RH, it was not possible to achieve the set RH, (described in more detail in the text below).

2: VPD = water vapour pressure deficit

3: The number of plants listed is the initial number of plants used, the number decreased in some cases due to various reasons, and in experiment 2, there were also two destructive studies, where additional plants were used as described further below.

(24)

There were more plants, than those used for measurements of growth and developmental parameters (as mentioned in Table 1). The extra plants where evenly distributed between the growth chambers and used for destructive measurements at the end of the winter bud

treatment. The number of plants used in destructive measurements is mentioned under biomass, and microscopy of buds.

During the experimental treatments, the VPD conditions were the same as during the pre- growing conditions. RH was adjusted to keep VPD at 0.5 kPa (Table 1). The irradiance conditions were also the same as during the pre-growing conditions, but the photoperiod varied (table 2), the length of the full light period and the day extension light period for the different experiments were as described in the following:

In experiment 1a and 1b (Table 1), half of the plants were transferred to a 12 h photoperiod, called short day (SD), for induction of bud set. These plants got full light as described above, followed by 12 h of darkness. For comparison, the other half of the plants were kept at a 24 h photoperiod, LD and got the same 12 h full light followed by 12 h low-intensity day extension light, i.e. the same light treatment as during the pre-growing treatment. The plants were distributed among three temperature treatments; 12°C, 18°C or 24°C and the temperatures were kept constant. The RH was adjusted to keep a VPD of 0.5 kPa (Table 1). In total experiment 1a and 1b had 6 different combinations of treatments each with 2 different light treatments and 3 different temperature treatments.

Experiment 2 and 3 (Table 1) started with 22 h photoperiod, and 2 h of darkness. In

experiment 2 the daylength was shortened by 2 h per week for 5 weeks, until SD was reached, i.e. a 12 h photoperiod and 12 h of darkness, followed by keeping the plants at 12 h

photoperiod for 4 weeks, i.e. the rest of the bud set treatment. The shortening of the

photoperiod was conducted during the day extension light period, i.e. the plants got 12 h full light + day extension light and the dark period was in the middle of the day extension light period. In experiment 2 there were two temperature treatments; 12°C or 18°C and the

temperature was kept constant like in experiment 1a and 1b, and the VPD was also here kept at 0.5 kPa. The light treatment was supposed be the same in experiment 3 as in experiment 2, but due to technical failure in daylength shortening the plants got 22 h light for 3 weeks instead of 1 week. This increased the period of shortening the daylength from 5 to 7 weeks, followed by 1 week with SD, i.e. 12 h light. Also, in the chamber with 18°C/ 12°C

(temperature treatments described below) the plants had a total of 4 days without

incandescent light, because a fuse went out twice during the period of 22 h light. Due to a lack

(25)

23

of access to growth chambers, the experiment could unfortunately not be repeated. Because of this, results from experiment 3 were emphasized less than the other results, and the

experiment was cut short. Experiment 3 also had two temperature treatments, here the treatments were alternated between two temperatures in each treatment; 24°C/18°C and 18°C/12°C. The temperature was increased and decreased gradually in the course of 3 h.

Including the time it took to change the temperatures, the 2 temperatures where kept for 12 h each, with the highest temperature during full light, and the coldest during the day extension light and the dark period. The chambers were not able to keep the set VPD of 0.5 kPa during the colder periods since the RH increases when it gets colder. The VPD therefore decreased during the colder period. Experiment 2 and 3 had 2 different temperature treatments and both temperature treatments had 2 different provenances, giving 4 different combinations of treatments.

Treatment for bud break

After the treatment for bud set under the SD, the plants were re-transferred to LD, i.e. the condition they got during the pre-growing period, 24h light,18°C, and 76% RH. The length of the treatments varied between the experiments (Table 1).

Growth parameters recorded during treatments

During the experimental treatments, several growth parameters were recorded to evaluate the development of the plants. The recordings carried out varied between the experiments, and the frequency of measurements and when the recordings were carried out varied (Table 2). The number of plants used in each case is shown in Table 1 and the same plants were used for all the recordings, except recordings of biomass and number of leaf primordia (table 2) where excess plants where used, the number of plants used in these 2 recordings is described below.

Height and cumulative growth

Height was measured from the rim of the pot to the shoot apical meristem. Afterwards cumulative growth was calculated by subtracting the height from the first measurement (cumulative growth during bud set) in experiment 2 and 3 and subtracting the height from the measurement conducted on the day of re-transfer to LD (cumulative growth during bud break). Due to a mistake in the first measurements (different persons measured in different ways) in experiment 1b and since the measurements from experiment 1a and 1b were

(26)

24

supposed to be compared with each other, it was decided to present height instead of growth in experiment 1a and 1b.

Bud set

During the SD conditions in experiment 1a and 1b and during the day shortening treatment in experiment 2 and 3, bud set development was recorded using a magnifier, and divided in to 3 categories: 0 = no visible bud and shoot elongation (Figure 2 A), 1 = white bud or light green bud (Figure 2 B and C), 2 = bud has turned brown (Figure 2 D and E).

Bud break

After the re-transfer to LD and 18°C (conditions as during pre-growing) bud break was registered, the registration was also done using a magnifier. The following 3 categories where used: 2 = intact bud (Figure 3 A), 1 = hole in bud (Figure 3 B and C), all needles were still gathered and bent inwards, 0 = needles were spreading away from each other (Figure 3 D).

Table 2 Overview of growth parameters recorded during the different experiments, studying the effect of temperature and daylength treatments on growth and development in Norway spruce. Day 0 was the day the treatments started.

Growth parameters

Experiments Frequency

Height 1b Once a week, 1^st,6^th and 7^th week missing 1a

Once a week 2

3

Bud formation 1a From day 12, twice a week 1b From day 1, twice a week

2 From day 0, twice a week until first bud, then 3 times a week.

3

Bud break 1a From day 50 (day 0 of LD), twice a week 1b

2 From day 65, (day 1 of LD), three times a week, once the last week.

Shoot diameter 2 Once a week, not the last week.

3 Once a week.

Number of needles

2 3 times: day 1, day 25 and day 63, all during SD treatment.

3 3 times: day 1, day 28 and day 56, all during SD treatment.

Number of lateral buds

2

Once, last day of SD (Day 63 and day 56, for experiment 2 and 3 respectively) 3

Stem diameter 2 3

Biomass 2

Number of leaf primordia

2

(27)

25

Figure 2 Pictures of bud development during temperature and daylength treatments, as examples of bud categories. Category 0: In picture A there is no visible bud. Category 1: In picture B there is a bud starting to be visible, and in picture C there is a white bud clearly visible. Category 2: In picture D the bud is starting to get brown and in picture E the bud has fully turned brown.

Figure 3 Pictures of bud break development after temperature and daylength treatments, after re- transfer to long day (LD) treatment. Category 2: In picture A the bud is fully closed. Category 1: In picture B there is a hole starting to form and in picture C the needles are starting to protrude out of the bud but are still gathered. Category 0: In picture D the needles have protruded fully from the bud and are no longer gathered.

(28)

26 Shoot diameter

Shoot diameter was measured from needle tip to needle tip across the top of the shoot with two, perpendicular measurements per plant. Then the average for the two measurements was calculated.

Needles

Needles longer than 5 mm were counted, but the needles that were closely gathered at the shoot tip were not counted, because this was not practically possible. Figure 2 A shows an example of needles gathered too much for counting.

Number of lateral buds

Lateral buds were counted with a distinction between closed and open buds. The total number of lateral buds was calculated afterwards.

Stem diameter

The thickness of the stem was measured right below the needles, before the stem thickened. A digital caliper (Digimatic, Pluss Mitutoyo Corporation, Kawasaki, Japan) was generally used, but during measurements in experiment 2, the digital caliper malfunctioned and a manual caliper was used for the rest of the measurements. During experiment 3 only the digital caliper was used.

Biomass

A total of 24 plants per provenance and temperature treatment were used for biomass recording. The plants were rinsed to remove soil and dried of with a paper towel. They were then separated into roots and shoots and were then dried in an oven at 70 °C for about a week, before the shoot was separated into needles and stem and weighed. The ratio between root weight and shoot weight was then calculated. Because this was a destructive study and it was conducted at the end of bud set treatment, excess plants were used, instead of the plant used in recordings described above, since they also were used for recordings after re-transfer to LD.

Microscopy of buds with leaf primordia

From each provenance and temperature treatment, 5 buds were collected. The buds were fixed in a 4% paraformaldehyde solution with 0.025% glutaraldehyde in a sodium phosphate buffer (PBS, 1M, pH 7.0). The PBS solution with the samples were placed in vacuum for 1 h to remove air within the buds. The solution was then replaced with new PBS buffer before the samples were stored at 4°C. Then the samples were washed with PBS buffer, and gradually dehydrated by placing the buds in solutions of ethanol with increasing concentrations (30%,

(29)

27

50%, 70%, 90% and 100%) and water. The samples were stored at 4°C in between each concentration of ethanol for at least 1 h. Then the samples were infiltrated with LR white resin (London Resin, Basingstoke, UK), by gradually increasing the ratio of LR white to ethanol, first 1:1 overnight, then 2:1 for 4 h, and then pure LR white for 3 days, where the LR white was replaced every 12 h. The samples were stored at 4°C under infiltration. Afterwards the buds were embedded in LR white by polymerization at 60°C. Then the samples were cut longitudinally into sections with a thickness of 1 µm on an ultramicrotome (Leica, Wetzlar, Germany), with a diamond knife (Diatome, Hatfield, PA, USA). The sections were placed on positively charged microscope slides (Superfrost, Thermo Fischer Waltham, MA, USA), that were laying on a hot plate at 55°C to adhere the samples to the slide glass. The samples were then stained with Stevenel’s blue (2% Potassium permanganate (KMnO4),1.3% methylene blue), then the samples were studied under a light microscope (Leica), to take micrographs and count the number of leaf primordia inside the bud. This was also a destructive study, so excess plants were used here as well.

Statistical analyses

A two-way analysis of variance (ANOVA glm) was used to analyse the data for all growth parameters in experiment 1a, 1b and 2, whereas a one-way ANOVA glm was used for experiment 3 since the daylength treatments in the two temperature regimes (in separate growth chambers) were not identical due to technical failure. The ANOVA analyses were followed by Tukey’s post hoc test. For lateral buds, the total number of lateral buds was analysed. For biomass, the root weight, stem weight, needle weight, shoot (stem and needle) weight, and the root/shoot ratio was analysed. For height (experiment 1a and 1b), cumulative growth during bud set (experiment 2 and 3), shoot diameter and number of needles the last registration of the bud set treatment was analysed. For cumulative growth during the bud break treatment, the last measurement of the bud break treatment was analysed. For bud data, both bud stages during bud set and bud break on individual days were analysed. The

significance level was set to p ≤ 0.05. Minitab 19 (Minitab Inc., State College, PA, USA) was used to conduct the analyses. The data were log transformed if the requirements for normal distribution or equal variance were not met.

The overall data (entire time course) for bud development in experiment 2 were also analysed with a cumulative link mixed model (clmm), with the same significance level as the rest of the data. R, version 3.5.1 (R Development Core Team 2019) was used to analyse the data.

(30)

28

Two packages were used: the ordinal package (Christensen 2019), for the clmm and the MuMIn package (Bartoń 2019) to compare models using Akaike information criterion corrected (AICc).

The bud data for both provenances in experiment 1a and 1b were analysed together, and days for analysis were chosen based on days where registration were conducted on the same day, while the height and cumulative growth data in experiment 1a and 1b were analysed

separately, because the last measurements were not conducted on the same day.

Since the results for experiment 3 where compromised due to technical failure, the two temperature treatments could not be compared directly since they did not experience identical daylength conditions. Therefore, the temperature treatments where analysed separately, in order to compare the responses of the provenances.

(31)

29

Results

Effects of constant daylength and temperature (experiment 1a and 1b) Growth during SD vs. LD

In both the Halden (from 59°N) and Rana (from 66°N) provenances, plants kept at different daylengths (SD of 12 h and LD of 24 h) and at different temperatures (12,18 or 24°C) had significantly different height at the end of the experiment, and there was a significant

interaction between daylength and temperature (all with a p-value < 0.001) (table 3, figure 4).

Both the Halden and Rana-plants that were exposed to SD ceased their growth during the treatment, and subsequently had grown significantly less than the plants that got LD treatment at the same temperature that continued to grow throughout the experiment. In the plants from Halden, growth cessation occurred after about 2 (24°C and 18°C) to 3 (12°C) weeks and growth slowed down before the growth cessation (figure 4 A). In the plants from Rana, the first measurement was missing due to a mistake. Therefore, it is not possible to say whether or how much the plants grew the first week. However, growth cessation occurred after about 2 weeks, in plants kept at 12°C and 18°C, while there was no growth in the plants grown at 24°C under the SD treatment (figure 4 B).

Figure 4 Effect of constant temperature (12,18 or 24°C ) under short days (SD) of 12 h photoperiod (bud set treatment) or long days (LD) of 24 h in on plant height seedlings of Norway spruce

provenances from Halden (A, from 59°N) and Rana (B, from 66°N). Different lowercase letters on the right side of the graphs indicate significant differences between treatments for the last measurement for each provenance separately, based on two-way ANOVA, followed by Tuckey`s test. N=20 per treatment. Average height with ± SE displayed.

For both provenances there was no significant difference between plants grown at 24°C and 18°C in the LD treatment, while plants kept at 12°C were significantly smaller and the height was about the same as plants from SD. In the SD treatment, the plants from the different provenances showed slightly different growth. The SD-exposed Halden-plants kept at 24°C were significantly taller than the plants grown at 12°C, while the Halden plants given 18°C were not significantly different from either of the other temperature treatments (figure 4 A). In

(32)

30

the SD-exposed Rana-plants, neither of the temperature treatments resulted in significantly different plant heights, but only the plants kept at 12°C and SD were significantly shorter than the plants exposed to 12°C and LD (figure 4 B).

The plants from Halden (figure 4 A) seemed to grow more than the plants from Rana (figure 4 B), in both daylength treatments and in all temperature treatments.

Table 3 ANOVA results for the final height of seedlings of the Halden (from 59°N, at day 49) and Rana (from 66°N, at day 48) provenances of Norway spruce exposed to constant daylength and temperature treatments (12, 18 or 24°C). The plants were either given 24 h daylength, long day or 12 h daylength, short day. N=20 per treatment.

Source DF Adj SS Adj MS F-Value P-Value Halden

Daylength 1 2.0038 2.00376 323.12 0.000 ***

Temperature 2 0.7212 0.36061 58.15 0.000 ***

Daylength*Temperature 2 0.1360 0.06799 10.96 0.000 ***

Error 111 0.6884 0.00620

Total 116 3.5101

Rana

Daylength 1 2.3913 2.39131 222.55 0.000 ***

Temperature 2 0.7373 0.36863 34.31 0.000 ***

Error 114 1.2249 0.01074

Total 119 4.9409

Significance codes: *** < 0.001, ** < 0.01, * ≤ 0.05

Bud set development

When analysing bud development under SD in Halden and Rana together, there was a significant effect of both temperature (12, 18 or 24°C) and provenance at all days analysed (day 15, 19, 22, 26, 40 and 47), and there was a significant interaction between temperature and provenance at all days except day 19 and day 26 (table 4). In both provenances, the plants kept at 24°C and 12°C had the fastest and slowest bud development, respectively. Plants from Halden (figure 5 A) had slower bud development than plants from Rana (figure 5 B).

(33)

31

Table 4 ANOVA table for bud set in seedlings of the Halden (from 59°N) and Rana (from 66°N) provenances of Norway spruce treated with constant temperatures (12, 18 or 24°C) under short days of a 12 h photoperiod. Individual days were analysed. N=20 per treatment.

Source DF Adj SS Adj MS F-Value P-Value Day 15

Temperature 2 8.017 4.0083 29.39 0.000 ***

Provenance 1 1.875 1.8750 13.75 0.000 ***

Temperature*Provenance 2 1.550 0.7750 5.68 0.004 **

Error 114 15.550 0.1364

Total 119 26.992

Day 19

Temperature 2 28.0167 14.0083 431.61 0.000 ***

Provenance 1 0.1333 0.1333 4.11 0.045 *

Temperature*Provenance 2 0.1167 0.0583 1.80 0.170

Error 114 3.7000 0.0325

Total 119 31.9667

Day 22

Temperature 2 43.398 21.6991 480.78 0.000 ***

Provenance 1 3.304 3.3043 73.21 0.000 ***

Temperature*Provenance 2 4.088 2.0440 45.29 0.000 ***

Error 113 5.100 0.0451

Total 118 55.966

Day 26

Temperature 2 35.3915 17.6957 111.97 0.000 ***

Provenance 1 5.4943 5.4943 34.77 0.000 ***

Temperature*Provenance 2 0.7604 0.3802 2.41 0.095

Error 113 17.8579 0.1580

Total 118 59.9328

Day 40

Provenance 1 3.5700 3.57000 417.13 0.000 ***

Temperature 2 5.9979 2.99896 350.40 0.000 ***

Error 111 0.9500 0.00856

Total 116 17.2308

Day 47

Temperature 2 1.5741 0.78704 17.47 0.000 ***

Provenance 1 0.8095 0.80952 17.97 0.000 ***

Error 111 5.0000 0.04505

Total 116 9.1453

(34)

32

Figure 5 Effect of constant temperature (12,18 or 24°C) under short day (SD) treatment of 12 h photoperiod on bud set in seedlings of Norway spruce provenances from Halden (A, from 59°N) and Rana (B, from 66°N). Different lowercase letters represent significant difference between treatments for the points on the left side of the letters for each provenance separately, based on one-way ANOVA followed by Tuckey`s test. When lines overlap and there are no significant differences, one letter is shown. 3 bud set categories were used: 0 = no visible bud and shoot elongation, 1 = white bud or light green bud, 2 = bud has turned brown. N=20 per treatment. Average bud set category with ± SE displayed.

Bud break development

In both provenances (Halden and Rana) there was a significant difference in time to bud break after the re-transfer to LD and 18°C between the plants from the different temperatures (12, 18 or 24°C) during the SD treatment of 12 h photoperiod at all days analysed. There was also a significant difference between the plants from different provenances on all days except day 14 and a significant interaction between temperature and provenance on all days analysed (table 5).

Figure 6 After-effect of constant temperature (12, 18 or 24°C) under short day (SD) treatment of 12 h photoperiod on bud break in seedlings of Norway spruce provenances from Halden (A, from 59°N) and Rana (B, from 66°N) after re-transfer to long days of 24 h photoperiod and 18°C (bud break treatment). Different lowercase letters represent significant difference between treatments for the points on the left side of the letters for each of the provenances separately, based on the one-way ANOVA followed by Tuckey`s test. When lines overlap and there are no significant differences, one letter is shown. 3 bud break categories where used: 2 = intact bud, 1 = hole in bud, all needles were still gathered and bent inwards, 0 = needles were spreading away from each other. N=19-20 per treatment. Average bud break ± SE displayed.

(35)

33

The plants from at 12°C had the fastest bud break in both provenances. The plants from 24°C were slowest in the Halden-plants, while there was no significant difference between the plants from 18°C and 24°C in the Rana- plants. The plants from Halden (figure 6 A) started to break buds earlier, while the plants from Rana had a faster development (figure 6 B), except the Rana- plants from 12°C, which were the fastest from the start.

Table 5 ANOVA table for bud break in seedlings of the Halden (from 59°N) and Rana (from 66°N) provenances of Norway spruce after re-transfer to long days and 18°C following exposure to different constant temperatures (12, 18 and 24°C) and short days of a 12 h photoperiod. Individual days were analysed. N=19-20 per treatment.

Source DF Adj SS Adj MS F-Value P-Value Day 11

Temperature 2 7.216 3.6082 25.40 0.000 ***

Provenance 1 1.859 1.8587 13.09 0.000 ***

Error 113 16.050 0.1420

Total 118 27.496

Day 14

Temperature 2 4.4595 2.22974 16.80 0.000 ***

Provenance 1 0.0330 0.03304 0.25 0.619

Error 113 15.0000 0.13274

Total 118 23.9664

Day 18

Temperature 2 5.6149 2.8074 14.59 0.000 ***

Provenance 1 0.9996 0.9996 5.19 0.025 *

Error 113 21.7500 0.1925

Total 118 33.9832

Day 21

Temperature 2 5.262 2.6310 20.94 0.000 ***

Provenance 1 5.584 5.5843 44.44 0.000 ***

Error 113 14.200 0.1257

Total 118 30.319

Day 25

Temperature 2 11.239 5.6196 45.07 0.000 ***

Provenances 1 3.960 3.9601 31.76 0.000 ***

Temperature*Provenance 2 1.989 0.9946 7.98 0.001 **

Error 113 14.089 0.1247

Total 118 31.143

(36)

34 Growth after re-transfer to LD and 18°C

After re-transfer to LD and 18°C from SD and LD at different temperatures (12, 18 and 24°C), plants exposed to the different daylengths and temperatures showed significantly different growth in both provenances (Halden and Rana), and there was a significant

interaction between temperature and daylength treatment (p-value < 0.001) (table 6). In both provenances, LD-exposed plants had grown significantly more than plants that got SD treatment. Halden-plants exposed continuously to LD at 18°C had grown significantly more than those transferred from LD and 12°C or 24°C, which were not significantly different from each other (figure 7 A). Rana-plants from LD treatment at 24°C and at 18°C did not differ significantly after transfer to 18°C but had grown significantly more than the plants from 12°C (figure 7 B). After the SD treatment, there was no significant difference between the temperature treatments for any of the provenances. Plants from Halden (figure 7 A) seemed to have grown slightly during the last week, while this was not the case for plants from Rana (figure 7 B).

Table 6 ANOVA results for cumulative growth at the final measurement day after re-transfer of seedlings of the Halden (from 59°N, at day 27) and Rana (from 66°N , at day 26) provenances of Norway spruce to long days (LD) of 24 h and 18°C from constant short days (SD) of 12 h photoperiod or 24 h LD under different constant temperatures (12,18 or 24°C). The data were log- transformed before analysis. N=19-20.

Source DF Adj SS Adj MS F-Value P-Value Halden

Daylength 1 9.0740 9.07403 525.98 0.000 ***

Temperature 2 0.6973 0.34865 20.21 0.000 ***

Error 109 1.8804 0.01725

Total 114 12.0887

Rana

Daylength 1 117.82 117.818 271.16 0.000 ***

Temperature 2 19.93 9.963 22.93 0.000 ***

Error 112 48.66 0.434

Total 117 207.63

Effect of temperature-photoperiod interaction on growth and winter bud development in Norway spruce (Picea Abies)

Effect of temperature-photoperiod interaction on growth and winter bud development in Norway spruce (Picea Abies)

Katharina Therese Hobrak

The Norwegian University of Life Sciences

Norges miljø- og biovitenskapelige universitet

Master thesis

Effect of temperature-photoperiod interaction on growth and winter bud development in Norway spruce (Picea Abies)

Katharina Therese Hobrak

Department of Plant Science Ås, 2020

The Norwegian University of Life Sciences

P.O Box 5003, 1432 Ås, Norway

Abstract

Sammendrag

Acknowledgements

Abbreviations

Keywords

Table of Contents

Introduction

Materials and methods

Results